Meredith Whitney predicted the 2007 financial crisis. At the time, economists and Wall Street analysts attacked her prediction. In 2010, she predicted a wave of municipal-bond defaults. As USA Today reports, the director of the National League of Cities called her prediction a “stunning lack of understanding.” Three cities in California declared bankruptcy within three weeks in 2013, including Stockton, the largest in US history. Those bankruptcies still did not faze the critics who called them isolated events.
Now Whitney is out with a book predicting the decline of coastal states and the rise of a corridor of Mid America including Colorado, the Dakotas, Indiana, Texas, and Utah. Whitney’s rationale for the prediction is that coastal states have overextended themselves with long-term pension obligations and will have to resort to raising already high taxes and cutting essential services to avoid bankruptcy. That, in turn, will drive people out of those states and into more fiscally conservative ones.
Whitney may be correct, if tax and services determine people’s choice of where to live, or she may be wrong if they do not. One does not need the Prospect Theory of the Nobel laureate Daniel Kahneman (and his late colleague, Amos Tversky), however, to predict that people with some interest in defending the municipal debt market, as well as those leaning to the left who defend the lavish policies of mostly liberal states and municipalities, will be unhappy with Whitney’s analysis despite her record of predictions. The director of the National League of Cities who called her ignorant in 2010 was clearly suffering from this bias.
More recently, Bloomberg’s Municipal Market Editor, Joe Mysac, has also been greatly unhappy with Whitney’s latest book, finding “factual errors.” (She missed Kansas’ bankruptcy in 1933!) Commensurate with the need to minimize her status as a good counterintuitive predictor, Mysac describes her 2007 prediction of the financial meltdown as limited to reporting that Citibank will suspend its dividend (she predicted much more than that,as Forbes and Fortune named her one of the top analysts in 2007 and 2008). An economist from California suggests sarcastically that “it is wonderful to be impervious to facts” when she declared California is already losing some of its wealthy entrepreneurs to lower-tax states (like Texas).
The facts, as usual, are ambiguous. On the one hand, jobs growth in 2012 and the first quarter of 2013 was faster in four states that Whitney named as laggards than in ten states she named as leaders. On the other hand, it is easy to grow when your numbers have been really bad for years prior to 2012.
But the more-interesting question, regardless of politics and economic models, is this: if a pundit predicted two things correctly, when very few others did, does it increase the probability that her third prediction will also be right? Does it decrease it? Or does predictive accuracy follow a binomial distribution of independent events, i.e., the probability that she is right remains 50% from one prediction to the next like the probability of heads when flipping a coin?
One way to answer this is to measure success over time. An analyst may pick the right stock based on a wild guess or a superior financial model. How do we know which is true? Kahneman and his colleagues showed in a famous study that, over time, star stock-picking performers fizzle out (zero correlation between one year’s performance and the next). Soros’ famous bet against the British pound in 1979 made him a financial guru, and a very rich man, but he has not repeated his maverick feat since then. And then of course there is the Uncertainty Principle: we can’t observe an electron’s position without influencing it – Buffett and Soros and others like them have such powerful images that their predictions become self-fulfilling prophecies as everyone follows then. This is no longer true clairvoyance.
A more-insightful analysis would look at the theory behind the prediction. Presently, there is no theory that can make consistently accurate prediction of stock-market movements. If anything, there is the opposite theory (random walk). But economic theory is not all guesses. Those predicting the housing-market collapse and the following financial collapse — and there were a few in addition to Whitney — all followed sound economic principles about the role of easy money in creating bubbles, government-imposed mortgage loans’ mandates, and banks’ incentive systems in creating distortions in lending. They were just being logical. Of course, logic alone will not let one pinpoint the exact date on economic consequences of certain policies; one needs additional signals. What makes Whitney and Steve Eisman (the famous “Big Short”) and a few like them unique was their ability to read weak, ambiguous signals ignored and derided by almost every pundit, administration official and other interested parties.. Of course, they may have just been lucky, but one does not get that impression upon reading their stories. The development of their foresight took time, persistence, and deep convictions about how economics work.
That last point has significant implications for competition and strategy. Reading weak, ambiguous signals constitutes the art of early warning. Whitney’s critics condemn her for ignoring mixed signs in 2012 and focusing instead on finding patterns of miniscule magnitude over several years, such as the net migration from California to Texas (only 0.1%.) The question is: when do you declare a pattern is forming? After everyone and their mother can see it?
This is a critical question for businesses, not just the investment community. With the huge hype around Big Data, decision makers are left to wonder: should we take the risk on weak signals or should we play it safe and wait until the marketing gurus declare it officially a trend and everyone jumps on the wagon? In this space of early warning, Whitney is a breath of fresh air. Yes, I like her politics, but that’s not the point. Whether or not she is right, she has already had an effect. Her warnings brought to focus the rating agencies’ complacency in grading debt and highlighted municipalities and states’ non-transparent accounting practices.
Quite often early warnings do not have to be accurate. They just have to have logic behind them.